We will continue our studies of plasticity in the auditory system using both experimental and theoretical methods. Experiments will examine the behavioral and electrophysiological consequences of several manipulations of organization rat auditory cortex. One series of experiments will use entirely physiological stimuli, and will force switches of attention between a pitch pattern and an intensity pattern in a discrimination task. A second series of experiments will directly manipulate cortex with the use of very weak electrical intra-cortical microstimulation (ICMS) in several different paradigms to modify the cortical sensory (auditory) map. Such electrical stimuli are far below direct behavioral/perceptual thresholds, but have been shown both in our and other labs to rapidly make changes in the relative size of cortical map domains (i.e. increase the number of neurons available for an appropriate task) and simultaneously to change the strength of the lateral connections between neurons in the cortical sheet. An ICMS period of minutes to several hours has effects on the cortex that last at least several hours,b ut then dissipate. For these experiments, the several stimulation paradigms will use either pure ICMS, or will combine ICMS with physiological stimuli (tone bursts), producing changes in the cortical network that are superficially similar in term of domain and receptive field changes, but which form a graded series of associations with "meaning." The effects of the cortical manipulations will be measured by performance of an appropriate pattern description task (pitch or pitch/intensity). The underlying cortical neuronal network will be studied by simultaneous separable recording of 30-50 neurons using stereotrode and tetrode technology Analysis of these data in terms of neuronal assembly properties and organization will use methods largely developed in our lab including the Joint PST histogram and the Gravity representation which delineate dynamic aspects of assembly interactions. Computer simulations will be used to mimic the observed experimental results and thus test likelihood of parameter values and mechanisms.